(this is a project together with the research team at the University of Twente in the Netherlands)

An INTERnet of SECure Things (INTERSECT) is critical to all aspects of the digital transformation: from energy to health, from smart industries to smart cities, from a sustainable environment to citizen trust and a resilient society. Yet, major breakthroughs are needed to achieve this. Not only technically (e.g. updating and managing devices) but also regarding security certification, effective IoT monitoring, and overcoming operational and societal challenges limiting governance, privacy, assurance, and crime prevention capabilities across all sectors. The INTERSECT project implements a truly interdisciplinary approach across Computer Science, Law and Criminology, and drives major breakthroughs in the IoT domains of Security-by-Design, Defence and Attack, Policy and Governance, and Privacy and Data Protection.

Our sub-component of the INTERSECT project investigates different ways to model the network behaviour of IoT devices in order to fingerprint IoT devices, but also to detect anomalies both regarding the devices themselves as well as the environment they interact with. Once an anomaly is detected, appropriate ML-based post-processing approaches will be explored to identify malicious anomalies and to deliver adequate courses of action to respond to the threats. Given the constantly evolving nature of attacks, the updating of used ML models will pose a particular challenge in this project. Furthermore, we will study manners to ease the interaction between human analysts and our to-be-developed detection tools to facilitate a fast response to attacks.